Antagonistic Epistasis of Hnf4α and Foxo1 Networks Through Enhancer Interactions

bioRxiv preprint doi: https://doi.org/10.1101/2020.07.04.187864; this version posted July 5, 2020. The copyright holder for this preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under aCC-BY-NC-ND 4.0 International license.

Kuo et al, Hnf4a and FoxO1 in b-cell function

Antagonistic epistasis of Hnf4α and FoxO1 networks through enhancer interactions

in β-cell function

Taiyi Kuo1,3,*, Wen Du1, Yasutaka Miyachi1, Prasanna K. Dadi2, David A. Jacobson2, Domenico

Accili1

1Department of Medicine and Berrie Diabetes Center, Columbia University College of

Physicians and Surgeons, New York, NY

2Department of Molecular Physiology and Biophysics, Vanderbilt University, Nashville, TN

3Lead Contact

*Correspondence: Taiyi Kuo, Department of Medicine and Berrie Diabetes Center, Columbia

University College of Physicians and Surgeons, 1150 St. Nicholas Avenue, Room 237, New

York, New York 10032, USA. Phone: 1-212-851-5333. Email: [email protected].

1 bioRxiv preprint doi: https://doi.org/10.1101/2020.07.04.187864; this version posted July 5, 2020. The copyright holder for this preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under aCC-BY-NC-ND 4.0 International license.

Kuo et al, Hnf4a and FoxO1 in b-cell function

Abstract

Genetic and acquired abnormalities contribute to pancreatic β-cell failure in diabetes.

Transcription factors Hnf4α (MODY1) and FoxO1 are respective examples of these two

components, and are known to act through β-cell-specific enhancers. However, their

relationship is unclear. Here we show by genome-wide interrogation of chromatin modifications

that FoxO1 ablation in mature β-cells leads to increased selection of FoxO1 enhancers by

Hnf4α. To model the functional significance we generated single and compound knockouts of

FoxO1 and Hnf4α in β-cells. Single knockout of either gene impaired insulin secretion in

mechanistically distinct fashions. Surprisingly, the defective β-cell secretory function of either

single mutant in hyperglycemic clamps and isolated islets treated with various secretagogues,

was completely reversed in double mutants. Gene expression analyses revealed the reversal of

β-cell dysfunction with an antagonistic network regulating glycolysis, including β-cell

“disallowed” genes; and that a synergistic network regulating protocadherins emerged as likely

mediators of the functional restoration of insulin secretion. The findings provide evidence of

antagonistic epistasis as a model of gene/environment interactions in the pathogenesis of β-cell

dysfunction.

2 bioRxiv preprint doi: https://doi.org/10.1101/2020.07.04.187864; this version posted July 5, 2020. The copyright holder for this preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under aCC-BY-NC-ND 4.0 International license.

Kuo et al, Hnf4a and FoxO1 in b-cell function

Introduction

Genetic predisposition contributes to type 2 diabetes through changes in the expression

and activity of transcription factors essential for β-cell function and insulin sensitivity (Ferrannini,

2010). The best characterized examples are the MODY gene mutations, which generally act in

autosomal dominant fashion to impair insulin secretion (Fajans and Bell, 2011). In the vast

majority of patients, however, such predisposition is more subtle, and requires acquired

abnormalities, often associated with altered nutrient utilization, to give rise to overt β-cell

dysfunction. An example of acquired transcriptional abnormalities leading to altered β-cell

function is FoxO1, a nutrient-regulated transcription factor whose nutrient excess-driven failure

leads first to metabolic inflexibility (Kim-Muller et al., 2014), and then to outright β-cell

dedifferentiation (Cinti et al., 2016; Sun et al., 2019; Talchai et al., 2012). This is achieved in

part by direct actions on glucose metabolism (Kuo et al., 2019b) and in part by regulating

lineage stability (Kuo et al., 2019a).

We have been interested in exploring the functional relation between FoxO1 and Hnf4α,

as a model of gene/environment interactions in the pathogenesis of β-cell dysfunction. When we

ablated FoxO1, 3a and 4 in pancreatic progenitors or mature β-cells, gene ontology analyses

revealed that Hnf4α targets were the most altered (Kim-Muller et al., 2014). However, whether

these changes resulted in activation or repression of Hnf4α, and whether they contributed to, or

offset FoxO1-induced β-cell dysfunction, was not clear.

Mutations of HNF4A cause MODY1 (Yamagata et al., 1996). Most patients develop

diabetes in the third decade of life owing to a primary defect in insulin secretion (Herman et al.,

1994) that can be treated long-term with sulfonylureas (Gardner and Tai, 2012). In mice,

pancreas-wide (Pdx1-Cre) or mature β-cell (Rip-Cre) deletion of Hnf4α causes glucose

intolerance due to defective insulin secretion (Boj et al., 2010; Gupta et al., 2005; Miura et al.,

2006). There are shared phenotypic features between FoxO-deficient and Hnf4α-deficient β-

3 bioRxiv preprint doi: https://doi.org/10.1101/2020.07.04.187864; this version posted July 5, 2020. The copyright holder for this preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under aCC-BY-NC-ND 4.0 International license.

Kuo et al, Hnf4a and FoxO1 in b-cell function

cells (Gupta et al., 2005; Kim-Muller et al., 2014; Miura et al., 2006) that include compromised

glucose-stimulated insulin secretion, and altered Pparα and Hnf1α gene network expression,

indicative of shared pathways and targets of both transcription factors.

The present studies were undertaken to analyze a potential FoxO1/Hnf4α epistasis.

Surveys of histone H3 lysine 27 acetylation for active promoters and enhancers uncovered an

interplay between Hnf4α and FoxO1 at β-cell enhancers. Thus, we generated and compared

mutant animals lacking FoxO1, Hnf4α, or both in β-cells. Surprisingly, FoxO1/Hnf4α double

knockout mice (DKO) showed restored glucose tolerance and insulin secretion compared to

single knockouts. These findings fundamentally change our understanding of the interactions

within transcriptional networks regulating β-cell function by highlighting a heretofore

unrecognized antagonistic epistasis (Snitkin and Segre, 2011).

4 bioRxiv preprint doi: https://doi.org/10.1101/2020.07.04.187864; this version posted July 5, 2020. The copyright holder for this preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under aCC-BY-NC-ND 4.0 International license.

Kuo et al, Hnf4a and FoxO1 in b-cell function

Results

Altered enhancer selection in β-cell failure

Activation of FoxO1 enhancers is a key event in β-cells in response to insulin-resistant

diabetes in db/db mice (Kuo et al., 2019b). We investigated FoxO1-regulated enhancers in a

multiparity model of diabetes associated with genetic ablation of FoxO1 (Kuo et al., 2019a;

Talchai et al., 2012). To this end, we subjected fluorescently sorted β-cells from multiparous

diabetic β-cell-specific FoxO1 knockout mice (DM) and glucose-tolerant controls (NGT) to

genome-wide chromatin immunoprecipitation with acetylated histone 3 lysine 27 antibodies

(H3K27ac) as a marker of active enhancers and promoters (Creyghton et al., 2010). DM mice

show evidence of multiparity-associated β-cell failure with random as well as re-fed

hyperglycemia as well as fasting and refed hypoinsulinemia (Kuo et al., 2019a; Talchai et al.,

2012). We observed a global increase of H3K27ac in active regions (Fig. S1A), and in individual

genes (Fig. S1B) of DM β-cells. Nearly half (44%) of H3K27ac peaks were located in introns,

14% in intergenic regions, and 16% in promoters (Fig. S1C). We analyzed the data based on

H3K27ac regions associated with promoters (defined as +/– 3kb from transcription start sites) or

distal enhancers (defined as beyond promoters) (Lenhard et al., 2012). Interestingly, we

detected significant changes in distal enhancers in the absence of FoxO1, regardless of age or

parity (Fig. 1A). In contrast, we found minimal alterations in the promoters (Fig. 1B). These data

confirm the importance of FoxO1 in distal enhancers (Kuo et al., 2019b).

To investigate the functional consequences of the altered distal enhancer profile of DM

β-cells, we mined transcription factor binding motifs in regions with increased or decreased

H3K27ac marks in DM vs. NGT. We reckoned that activation of transcription factors would be

associated with enrichment of the relevant motifs in hyper-acetylated enhancers, whereas

suppression of transcription factors would be associated with enrichment of their motifs in hypo-

acetylated enhancers (Kuo et al., 2019b). FoxO1 was the second most-underrepresented motif

5 bioRxiv preprint doi: https://doi.org/10.1101/2020.07.04.187864; this version posted July 5, 2020. The copyright holder for this preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under aCC-BY-NC-ND 4.0 International license.

Kuo et al, Hnf4a and FoxO1 in b-cell function

found within hypo-acetylated enhancers in DM β-cells, thus validating the analysis (Fig. 1C). We

also found a significant depletion of the Pax6 motif, a FoxO1 target (Kim-Muller et al., 2016b;

Kitamura et al., 2009) that regulates Ins1, Ins2 (Sander et al., 1997), Nkx6.1, Pdx1, Slc2a2 and

Pc1/3 in β-cells (Hart et al., 2013). Pax6 represses alternative islet cell genes including ghrelin,

glucagon, and somatostatin (Swisa et al., 2017), and ablation of Pax6 in the adult pancreas

leads to glucose intolerance due to β-cell dedifferentiation (Hart et al., 2013; Swisa et al., 2017).

The data are consistent with the possibility that combined abnormalities of FoxO1 and Pax6

contribute to β-cell dedifferentiation.

In contrast, Hnf4α was the top hyper-acetylated motif with increased H3K27ac in DM β-

cells (Fig. 1C), consistent with prior gene ontology analyses indicating activation of the Hnf4α

network in triple FoxO (Kim-Muller et al., 2014) and single FoxO1 knockout β-cells (Kuo et al.,

2019a). In addition, Hic (Liu et al., 2011) and Ets (Luo et al., 2014), two networks of potential

interest for β-cell stability, were over-represented. Interestingly, Ets1 overexpression in β-cells

suppresses insulin secretion (Chen et al., 2016). These data reveal a striking rearrangement of

active enhancers in diabetic β-cells.

Restoration of metabolic parameters in FoxO1 and Hnf4α double knockout mice

To examine whether the increased representation of Hnf4α enhancers is compensatory

or contributory vis-à-vis β-cell dysfunction, we generated β-cell-specific FoxO1 (IKO), Hnf4α

(HKO) (Miura et al., 2006) as well as double knockout (DKO) mice using Rip-Cre transgenic

mice (Herrera, 2000) that do not give rise to extra-pancreatic recombination or produce Gh

mRNA (Brouwers et al., 2014). The animals were born to term in the expected Mendelian ratios,

and showed no growth abnormalities (Fig. S2A).

We performed oral glucose tolerance tests. As previously reported, IKO or HKO mice

displayed glucose intolerance (Fig. 2A) (Kuo et al., 2019a; Miura et al., 2006; Talchai et al.,

2012). Unexpectedly, compared to single knockouts, DKO mice showed restoration of normal

6 bioRxiv preprint doi: https://doi.org/10.1101/2020.07.04.187864; this version posted July 5, 2020. The copyright holder for this preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under aCC-BY-NC-ND 4.0 International license.

Kuo et al, Hnf4a and FoxO1 in b-cell function

glucose tolerance (Fig. 2A and 2B). We also assessed plasma glucose levels within 5 minutes

of intraperitoneal glucose injection as a surrogate measure of insulin secretion. Consistent with

glucose tolerance tests, IKO mice showed elevated glucose levels within 5 minutes, whereas

DKO mice showed no difference compared to WT (Fig. 2C). To investigate the nutrient

response, we measured glucose and insulin levels in fasted and refed mice. While no difference

was found in fasted plasma glucose and insulin levels between all 4 genotypes, refeeding led to

elevated plasma glucose and reduced insulin levels in IKO mice compared to WT (Fig. 2D, 2E).

Interestingly, DKO and WT mice show similar levels of glucose and insulin in response to a

meal (Fig. 2D, 2E).

Next, to assess insulin secretory capacity in vivo, we performed hyperglycemic clamps.

By intravenous glucose administration, we raised glycemia to ~350 mg/dL (Fig. S2B and S2C),

and measured the rate of glucose infusion necessary to maintain this level. DKO showed

increased glucose infusion rates compared to IKO and HKO, to a level indistinguishable from

WT (Fig. 2F and 2G). Plasma insulin levels increased accordingly, consistent with an improved

insulin secretory capacity in DKO mice (Fig. 2H and 2I). In contrast, all 4 genotypes responded

to exogenous insulin injection to a comparable extent (Fig. S2D). These data are consistent with

the possibility that FoxO1 and Hnf4α have antagonistic epistasis in β-cells, and that increased

Hnf4α enhancer occupancy contributes to β-cell dysfunction in the absence of FoxO1.

Secretagogue-stimulated insulin secretion and calcium flux

To investigate the mechanism of restored insulin secretion, we subjected purified islets

from all four genotypes to insulin secretion in response to various secretagogues. Consistent

with the in vivo studies, IKO and HKO islets showed impaired insulin secretion in response to

high glucose or arginine, which was restored to WT levels in DKO islets (Fig. 3A and 3B).

Similar to studies in triple FoxO knockout islets, IKO islets also showed compromised insulin

secretion in response to the sulfonylurea glibenclamide and the PKC activator PMA (Kim-Muller

7 bioRxiv preprint doi: https://doi.org/10.1101/2020.07.04.187864; this version posted July 5, 2020. The copyright holder for this preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under aCC-BY-NC-ND 4.0 International license.

Kuo et al, Hnf4a and FoxO1 in b-cell function

et al., 2014). Both responses were normal in HKO, and were restored in DKO islets (Fig. 3C and

3D). Although the sulfonylurea response differs from previously reported data in Hnf4α knockout

β-cells (Gupta et al., 2005; Miura et al., 2006), it is indeed consistent with the clinical response

of MODY1 patients (Gardner and Tai, 2012).

To probe the amplifying signal in insulin secretion, we tested exendin-4, a GLP-1

agonist, and GIP treatment. We found that both exendin-4- and GIP-stimulated insulin secretion

were comparable in WT, IKO, and DKO islets (Fig. 3E and 3F), indicating that the GLP-1 and

GIP receptor-dependent pathways of insulin secretion were not affected by FoxO1 ablation.

Furthermore, it has been shown that GLP-1 can stimulate insulin secretion in HKO islets,

although the magnitude was trending lower compared to WT (Miura et al., 2006).

Glycolysis-generated ATP binds to and closes KATP channels resulting in membrane

depolarization and activation of voltage-dependent calcium channels (VDCCs). Calcium influx

through VDCCs induces fusion of insulin granules to the β-cell plasma membrane and insulin

exocytosis. We investigated calcium flux in islets of various genotypes. Unlike triple FoxO

knockout islets (Kim-Muller et al., 2014), membrane depolarization-induced calcium influx in

response to glucose and KCl was slightly increased in IKO (Fig. 4A, 4B). In addition to the

genotype, differences in experimental design and time course of measures may account for this

difference. Consistent with prior observations, HKO islets showed decreased magnitude and a

delayed onset of calcium influx when compared to WT islets (Fig. 4A, 4B). Interestingly, DKO

restored the magnitude but not the delayed onset of calcium flux. As a result, the area under the

curve of the calcium responses following glucose-stimulation was restored by DKO at later, but

not earlier time points (Fig. 4A, 4B). In sum, FoxO1 and Hnf4α affect insulin secretion in

mechanistically distinct ways and result in antagonistic epistatic interactions.

Different mechanisms of β-cell dysfunction in FoxO1 vs. Hnf4α mutants

8 bioRxiv preprint doi: https://doi.org/10.1101/2020.07.04.187864; this version posted July 5, 2020. The copyright holder for this preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under aCC-BY-NC-ND 4.0 International license.

Kuo et al, Hnf4a and FoxO1 in b-cell function

To understand the mechanisms of epistasis between FoxO1 and Hnf4α, we introduced a

fluorescent ROSA26-Tomato reporter in the various mutant strains and used it to sort

genetically-labeled Tomato-positive β-cells (Kuo et al., 2019a). We performed RNAseq in WT,

IKO, HKO, and DKO β-cells and identified expression changes by differential gene-calling. Our

working hypothesis was that functional antagonistic epistasis could result from: (i) synergistic

regulation of networks that inhibit insulin secretion; (ii) antagonistic regulation of networks that

promote insulin secretion; (iii) mixed regulation of distinct components of individual networks

(Fig. 5) (Snitkin and Segre, 2011).

When compared to WT, there were changes to 769 mRNAs in IKO β-cells (Table S1),

1,327 in HKO (Table S2), and 1,692 in DKO (Table S3). Thus, the restoration of insulin

secretion in DKO seemingly is not due to a normalization of gene expression, but rather to the

emergence of a distinct gene expression profile. This is confirmed by comparisons of single with

double mutants, where DKO β-cells displayed 1,787 mRNA changes compared to IKO (Table

S4), and 666 compared to HKO (Table S5). Furthermore, there were 1,165 mRNA changes

between IKO and HKO β-cells (Table S6).

In IKO β-cells, we detected the expected alterations of known FoxO1 target mRNAs,

with increased glycolytic genes (Buteau et al., 2007), including β-cell “disallowed” genes (Kuo et

al., 2019b) and Aldh1a3 (Kim-Muller et al., 2016a), as well as decreased Cyb5r3 (Fan et al.,

2020) (Table S1). These data are consistent with previous findings indicating that triple FoxO

ablation results in altered metabolic coupling (Kim-Muller et al., 2014). Specifically, the

combination of increased glycolysis and impaired coupling to oxidative phosphorylation by

activation of disallowed genes may result in lower efficiency of ATP production and increased

ROS generation (Robertson et al., 2003).

Previous reports differ with regard to gene expression patterns associated with Hnf4α

deletion in β-cells (Gupta et al., 2005; Miura et al., 2006). In our cross, the most striking gene

expression changes in HKO were the substantial increases of abundant mRNAs encoding

9 bioRxiv preprint doi: https://doi.org/10.1101/2020.07.04.187864; this version posted July 5, 2020. The copyright holder for this preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under aCC-BY-NC-ND 4.0 International license.

Kuo et al, Hnf4a and FoxO1 in b-cell function

mitochondrial cytochrome components (COX, CYTB, ND1, ND2, ND4, and ND5) (Table S2).

Conversely, the mitochondrial complex III oxidoreductase Cyb5r3 was decreased, possibly

resulting in altered electron transport across the mitochondrial chain (Fan et al., 2020). There

were extensive increases in the family of KCN-type potassium channels, which may affect

membrane depolarization and insulin secretion, and thus explain the preserved response to

sulfonylurea (Fig. S3A). The levels of L-type VDCC Cacnb3 mRNA were decreased, a finding

that may partly explain the impaired calcium response in HKO islets (Table S2). Interestingly,

this change was not reversed in DKO, consistent with the observation that first-phase calcium

flux is NOT restored, despite the overall improvement of calcium flux and insulin secretion

(Table S3). These data indicate that the underlying mechanisms of islet dysfunction differ in IKO

vs. HKO islets. There were two notable exceptions to this: Cyb5r3, as indicated above, and

Gpr119, an orphan GPCR that is mechanistically linked to insulin secretion (Abdel-Magid,

2019), were equally decreased in both IKO and HKO (Table S1 and S2).

Epistatic network analysis

Clues as to the potential mechanism of reversal of β-cell dysfunction emerged from

comparing single mutants with DKO as well as WT (Table S3). The glycolytic pathway, including

key β-cell “disallowed” genes, such as Ldha and Slc16a3, was substantially decreased in DKO

and WT compared to IKO. Indeed, DKO levels were even lower than WT islets. The predicted

outcome of this change would be a decrease in glucose utilization and tighter coupling of

glycolysis with oxidative phosphorylation, owing to reduced Ldha activity and monocarboxylate

transport. This change is consistent with synergistic antagonism of FoxO1 and Hnf4α on the

glycolytic cascade, with the former acting as a suppressor and the latter as an activator of this

pathway (Fig. 5A, 5B).

In contrast, changes to HKO gene expression patterns were not reversed in either WT or

DKO islets. For example, the patterns of KCN expression remained virtually identical (Fig. S3A).

10 bioRxiv preprint doi: https://doi.org/10.1101/2020.07.04.187864; this version posted July 5, 2020. The copyright holder for this preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under aCC-BY-NC-ND 4.0 International license.

Kuo et al, Hnf4a and FoxO1 in b-cell function

Even when analyzing the vast and heterogenous family of solute carriers (Slc), 61% of the

changes seen in HKO failed to be reversed, as opposed to 23% in IKO islets. Thus, the majority

of Hnf4α targets are not epistatic with FoxO1. However, a remarkable synergistic regulation of

the protocadherin gene family emerged by comparing HKO with WT or DKO islets (Fig. 5C).

Protocadherins are thought to participate in the formation of adherens (tight) junctions between

β-cells that facilitate calcium influx and fusion of secretory vesicles (Dissanayake et al., 2018).

This may explain the unique patterns of calcium entry in DKO islets, which are slower but reach

higher levels than WT. This regulation of the protocadherin family is consistent with a model of

synergistic epistasis. In this instance, both FoxO1 and Hnf4α act in a similar fashion to repress

expression of these genes, and it’s only when both are removed that the full extent of the

regulation becomes apparent.

A second example of synergistic epistasis between FoxO1 and Hnf4α were mRNAs

encoding non-β-cell hormones (Fig. S3B). Gcg, Sst, Npy, Ppy, and Pyy increased substantially

in DKO islets compared to IKO, consistent with reduced lineage stability, a feature of “resting” or

dedifferentiated β-cells (Buteau et al., 2007; Talchai et al., 2012; Talchai and Accili, 2015).

Relevance to human type 2 diabetes GWAS

To investigate the human disease relevance of our findings, we compared DKO with

either single IKO mutants or WT. We integrated enhancer occupancy and RNAseq data to

select genes with increased H3K27ac marks and mRNA levels, or decreased H3K27ac and

mRNA levels. We selected genes concordantly up- or down-regulated in DKO compared to both

WT and IKO as potential “compensatory” genes, and queried a database of genetic variations in

human type 2 diabetes GWASdb SNP (Rouillard et al., 2016) (URL:

http://amp.pharm.mssm.edu/Harmonizome) to identify those associated with type 2 diabetes

(Tables S7 and S8).

11 bioRxiv preprint doi: https://doi.org/10.1101/2020.07.04.187864; this version posted July 5, 2020. The copyright holder for this preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under aCC-BY-NC-ND 4.0 International license.

Kuo et al, Hnf4a and FoxO1 in b-cell function

Genes with increased mRNA and H3K27ac numbered 12 in the DKO vs. WT, and 7 in

the DKO vs. IKO comparison. Two (Svil and Prex1) were shared in common. Genes with

decreased mRNA and H3K27ac were 22 in DKO vs. WT, and 32 in DKO vs. IKO β-cells

(Tables S7 and S8, Fig. S4). The latter included ARHGEF9, one of only 20 genes concordantly

decreased in three independent RNA profiling datasets of type 2 diabetes patients (Zhong et al.,

2019). 12 genes (Ttc22, Sel1l3, Rspo4, Pla2g4b, Nr1h4, Lad1, Jmjd7, Fut1, Elovl2, Dll4,

Cacnb3, and Baiap2l2) were shared in common between the two comparisons (Table S7 and

S8). Altogether, three genes were associated with human variant sequences conferring an

increased risk of type 2 diabetes: Svil, Prex1, and Elovl2.

12 bioRxiv preprint doi: https://doi.org/10.1101/2020.07.04.187864; this version posted July 5, 2020. The copyright holder for this preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under aCC-BY-NC-ND 4.0 International license.

Kuo et al, Hnf4a and FoxO1 in b-cell function

Discussion

β-cell function is maintained through a restricted network of transcription factors (Fujitani,

2017). However, the redundancy, heterogeneity, and complexity of gene expression patterns

associated with experimental or naturally occurring perturbations in the activity of these “master”

regulators complicate the task of interrogating the mechanisms of dysfunction resulting from

their interactions. The main findings of this work are: i) FoxO1 deletion results in a redistribution

of Hnf4α to FoxO1 distal enhancers in β-cells; ii) this redistribution underlies an antagonistic

epistasis between these transcription factors affecting insulin secretion; iii) gene expression

analyses reveal different epistatic modalities regulating different networks, examples of which

include antagonistic epistasis of the glycolysis pathway, and synergistic epistasis of the

protocadherin gene family. Epistatic synergism between paralogues in β-cell function has been

proposed (Boj et al., 2010). But antagonistic epistasis between distinct factors through cis-acting

enhancers and resulting in mechanistically distinct phenotypes had not been disclosed.

The interaction between FoxO1 and Hnf4α occurs at multiple levels. The two proteins

physically interact in liver (Puigserver et al., 2003). In addition, Hnf4α-initiated glucokinase

transcription is repressed by FoxO1, acting as a co-repressor (Ganjam et al., 2009; Hirota et al.,

2008) by way of SIN3a (Langlet et al., 2017). However, the co-regulation of pancreatic β-cell

distal enhancers by FoxO1 and Hnf4α has not been revealed until now, and its relevance to the

progression of common forms of diabetes in humans will have to be further explored.

Type 2 diabetes is a polygenic disease, and β-cell failure has acquired and genetic

components. FoxO1 can be considered an example of the former, and Hnf4α of the latter. Both

homozygous null mutations in β-cells are associated with an insulin secretory defect, although

the mechanisms appear to differ. Thus, the absent sulfonylurea response in FoxO1 knockout

mice is reminiscent of the acquired failure seen in diabetic patients, while Hnf4α knockout show

a preserved sulfonylurea response, like MODY1 patients (Gardner and Tai, 2012). Likewise, the

13 bioRxiv preprint doi: https://doi.org/10.1101/2020.07.04.187864; this version posted July 5, 2020. The copyright holder for this preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under aCC-BY-NC-ND 4.0 International license.

Kuo et al, Hnf4a and FoxO1 in b-cell function

calcium response to glucose differs in the two mutants, although this may partly be explained by

the compensatory function of FoxO3 in single FoxO1 knockouts (Kim-Muller et al., 2014).

Evidence of antagonistic epistasis between these factors arises from the surprising

phenotype of the double mutant mice. However, the majority of Hnf4α-dependent changes are

unaffected by the FoxO1 mutation, indicating that epistasis between these two proteins occurs

primarily within the FoxO1 regulome. For example, the Hnf1α network is inhibited in HKO

mutants (data not shown), and this alteration is not rescued in DKO mice. Conversely, most

changes induced by FoxO1 ablation appear to be offset by the Hnf4α mutation. The most

striking example is the antagonistic regulation of glycolysis, where FoxO1 acts as a suppressor

and Hnf4α as an activator of gene expression. Although the association of reduced expression

of glycolytic genes with restored insulin secretion in DKO islets may appear counterintuitive, it

includes suppression of Lhda, and therefore can be conducive to better metabolic coupling of

glycolysis with mitochondrial oxidative phosphorylation, leading to β-cell “rest” and reduced

ROS production (Robertson et al., 2003).

A second example of epistatic network is represented by the synergistic suppression of

protocadherins. The release of this suppression in DKO mice may lead to tighter coupling of

neighboring β-cells and therefore to faster propagation of action potentials, reducing

heterogeneity among β-cells (Dissanayake et al., 2018). Consistent with this hypothesis is the

observation that calcium influx into DKO occurs even more rapidly than in WT β-cells. While

further studies will be required to investigate this mechanism, these observations raise the

possibility that agents that improve cell-cell communication can restore insulin secretion in

diabetes.

An additional mechanism connecting the functional restoration of insulin secretion to

changes in intracellular vesicle trafficking emerges from the integration of RNAseq with

enhancer occupancy and human diabetes GWAS or RNAseq data. We found alterations of

Arhgef9, encoding Rho guanine nucleotide exchange factor 9, a guanine exchange factor for

14 bioRxiv preprint doi: https://doi.org/10.1101/2020.07.04.187864; this version posted July 5, 2020. The copyright holder for this preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under aCC-BY-NC-ND 4.0 International license.

Kuo et al, Hnf4a and FoxO1 in b-cell function

Cdc42. This is one of the few genes consistently identified in multiple surveys of diabetic islet

gene expression (Zhong et al., 2019). Small G-proteins, such as Rho, Rac1, Cdc42, and Arf6

play important roles in cytoskeletal remodeling, vesicle fusion, and insulin exocytosis (Kowluru,

2017). They cycle between inactive (GDP-bound) and active (GTP-bound) conformations,

regulated by a combination of activating guanine nucleotide exchange factors (GEFs),

inactivating GTPase-activating proteins (GAPs), and inhibitory GDP-dissociation inhibitors

(GDI). In particular, Cdc42 depletion leads to a loss of second-phase insulin secretion, while

Cdc42 activation is selectively responsive to glucose but not KCl (Wang et al., 2007). Thus,

Arhgef9/Cdc42-dependent cytoskeletal rearrangement can be a mechanism to rescue the

secretory phenotype in DKO vs. IKO β-cells.

Two of the genes associated with diabetes susceptibility in human GWAS studies and

dysregulated in IKO or DKO islets are also related to cytoskeletal remodeling. Prex1

(phosphatidylinositol-3,4,5-triphosphate-dependent Rac exchange factor 1) activates Rac1,

leading to cytoskeletal rearrangement (Balamatsias et al., 2011). β-cell-specific Rac1 knockout

mice display reduced glucose-stimulated insulin secretion due to failure to depolymerize F-actin,

which prevents recruitment of insulin granules and insulin secretion (Asahara et al., 2013). Svil

(encoding Supervillin) bridges membranes with the actin cytoskeleton (Chen et al., 2003). The

combined increases in Arhgef9-Cdc42 and Prex1-Rac1 provide a potential link between the Rho

GTPase family and cytoskeletal remodeling-mediated insulin secretion, and thus a

compensatory mechanism to explain the pattern of calcium influx in DKO β-cells. F-actin forms

a web beneath the plasma membrane that is often viewed as a barrier to the release of insulin

granules in basal conditions (Kalwat and Thurmond, 2013; Orci et al., 1972). The

depolymerization of F-actin is essential for nutrient-stimulated, second phase insulin secretion,

which requires recruitment of insulin granules from the intracellular storage pools to the plasma

membrane. The cyclic nature of F-actin remodeling, regulated by Cdc42 and Rac1, is necessary

for the regulation of insulin granule exocytosis (Kalwat and Thurmond, 2013).

15 bioRxiv preprint doi: https://doi.org/10.1101/2020.07.04.187864; this version posted July 5, 2020. The copyright holder for this preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under aCC-BY-NC-ND 4.0 International license.

Kuo et al, Hnf4a and FoxO1 in b-cell function

In summary, increased occupancy of Hnf4α through its re-distribution to FoxO1 distal

enhancers is contributory to β-cell dysfunction, highlighting the role of antagonistic epistasis

between transcription factors in the maintenance of pancreatic β-cells. These findings can

provide a model to understand the pathophysiology of β-cell dysfunction.

16 bioRxiv preprint doi: https://doi.org/10.1101/2020.07.04.187864; this version posted July 5, 2020. The copyright holder for this preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under aCC-BY-NC-ND 4.0 International license.

Kuo et al, Hnf4a and FoxO1 in b-cell function

Methods

Animals

We obtained Hnf4α flox/flox mice from Dr. Frank J. Gonzalez (Miura et al., 2006), and crossed

Rip-Cre/Rosa26-Tomato/floxed FoxO1 mice with floxed Hnf4α (Kuo et al., 2019a) to generate β-

cell-specific FoxO1 and Hnf4α knockout mice. We categorize mice with at least one allele of

FoxO1 and Hnf4α as WT for DKO studies. All mice were fed chow diet, and maintained on a 12-

h light cycle (lights on at 7am). Epigenetic screening was performed in female mice, and FoxO1

and Hnf4α double knockout validations were carried in male mice.

Metabolic Parameters

We performed intraperitoneal glucose tolerance tests with glucose (2g/kg) after an overnight (16

h) fast, and insulin tolerance tests by injecting insulin (0.75 units/kg) after a 5-h fast (Kuo et al.,

2016). For hyperglycemic clamps, we followed a previously described protocol (Kuo et al.,

2019b). To assess insulin capacity in vivo, we raised the plasma glucose level to ~350 mg/dL by

continuous infusion of glucose, and maintained by adjusting the glucose infusion rate on the

pump. Throughout the 120 min period, we recorded glucose infusion rates and plasma glucose

levels, and collected plasma for insulin measurement.

Fluorescence-assisted β-cell sorting

Pancreatic islets were isolated as described and allowed to recover in RPMI supplemented with

15% FBS overnight (Kuo et al., 2016). The next day, islets were washed with PBS, and

dissociated with trypsin, followed by quenching with FBS. Thereafter, the dispersed islets were

spun down and washed with PBS, and incubated with sytox red (Thermo Fisher, S34859) to

identify live cells and DNase I (Sigma, D4513) on ice until sorting. Prior to sorting, cells were

filtered through a 35-μm cell strainer (Corning 352235) to remove clusters. Tomato-positive and

17 bioRxiv preprint doi: https://doi.org/10.1101/2020.07.04.187864; this version posted July 5, 2020. The copyright holder for this preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under aCC-BY-NC-ND 4.0 International license.

Kuo et al, Hnf4a and FoxO1 in b-cell function

-negative cells differed by ~2 orders of magnitude in tomato fluorescence with excitation of

554nm and emission of 581nm by InfluxTM cell sorter (BD Biosciences).

H3K27ac ChIPseq

Islet β-cells were genetically labeled with ROSA26-Tomato fluorescence with Rip-Cre allele, and

FAC-sorted β-cells were used for histone H3K27ac ChIPseq with anti-H3K27ac antibody (Active

Motif, 39133). ChIP and ChIPseq were performed as previously described (Kuo et al., 2019a;

Kuo et al., 2012). The resulting DNA libraries were quantified with Bioanalyzer (Agilent), and

sequenced on Illumina NextSeq 500 with 75-nt reads and single end. Reads were aligned to

mouse genome mm10 using the Burrows-Wheeler Aligner (BWA) algorithm with default settings

(Li and Durbin, 2009). These reads passed Illumina’s purity filter, and aligned with no more than

2 mismatches. Duplicate reads were removed, and only uniquely mapped reads with a mapping

quality greater or equal to 25 were subjected for further analysis. Alignments were extended in

silico at their 3’ ends to a length of 200-bp, which is the average genomic fragment length in the

size-selected library, and assigned to 32-nt bins along the genome. Peak locations were

determined using the MACS algorithm (v1.4.2) with a cutoff of p value = 1×10-7 (Zhang et al.,

2008). ROSE was used to identify enhancers (Whyte et al., 2013). To generate scatterplots of

H3K27ac in promoter or distal enhancer regions, the parameters for differential peak calling

applied are the following: FPKTM >10 for at least one of the two comparing groups, and FC

>1.5.

Enhancer motif analysis

For H3K27ac ChIPseq, HOMER was used to create tag directories and to call broad peaks.

Cisgenome was used to find the distribution of H3K27ac peaks in the genome. We used a cutoff

of +/– 3 kb from the TSS to separate enhancers and promoters. De novo motif analysis was

done using HOMER in the troughs of differentially acetylated enhancers identified from

18 bioRxiv preprint doi: https://doi.org/10.1101/2020.07.04.187864; this version posted July 5, 2020. The copyright holder for this preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under aCC-BY-NC-ND 4.0 International license.

Kuo et al, Hnf4a and FoxO1 in b-cell function

H3K27ac. To compare peak metrics between samples, common intervals are categorized as

“Active Regions”, defined by the start coordinate of the most upstream interval and the end

coordinate of the most downstream interval. Active regions also include those intervals found in

only one sample. Tag density represents DNA fragment per 32-bp bin. The parameters for

differential peak calling are the following: Fragments Per Kilobase of DNA per Ten Million

mapped reads (FPKTM) >30 for at least one of the groups, and Fold Change (FC) >1.5.

RNAseq

We isolated total RNA Nucleospin RNA kit (Macherey-Nagel) with DNase I treatment, and

followed previously described protocol (Kuo et al., 2014). Directional Poly-A RNAseq libraries

were prepared and sequenced as PE42 (42-bp paired-end reads) on Illumina NextSeq 500. For

the analysis, 29 to 45 million read-pairs per sample were used. The reads were mapped to the

mouse genome mm10 using STAR (v2.5.2b) (Dobin et al., 2013) algorithm with default settings.

FeatureCounts from the Subread (v1.5.2) (Liao et al., 2014) package was used to assign

concordantly aligned reads pairs to RefSeq genes. A 25-bp minimum as overlapping bases in a

fragment was specified for read assignment. Raw read counts were used as input for DESeq2

(v1.14.1) (Love et al., 2014), which was further used to filter out genes with low counts,

normalize the counts using library sizes, and perform statistical tests to find significant

differential genes. Differential calling results were presented in Supplemental Table S1-6. For

statistical analysis, a cutoff of p value < 0.05 and FDR < 0.1 was applied, and the resulting

genes were used as input for Gene ontology analyses (Ingenuity Pathway Analysis software,

Qiagen). Raw and processed data were deposited into the MINSEQE-compliant National Center

for Biotechnology Information Gene Expression Omnibus database.

Fura-2 AM imaging of whole islet secretagogue-stimulate calcium influx

19 bioRxiv preprint doi: https://doi.org/10.1101/2020.07.04.187864; this version posted July 5, 2020. The copyright holder for this preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under aCC-BY-NC-ND 4.0 International license.

Kuo et al, Hnf4a and FoxO1 in b-cell function

Purified islets were loaded with 2 μM Fura-2-acetoxymethyl ester (AM) for 25 min at 37°C with

5% CO2. Islets were washed twice with Kreb’s buffer, and incubated in Kreb’s buffer

supplemented with 2 mM glucose for 20 min to reach basal condition before imaging. Then, 14

mM glucose and 45 mM KCl were added sequentially to stimulate calcium flux. Relative

intracellular calcium concentrations were measured every 5 sec as a ratio of Fura-2

fluorescence excitation at 340 and 380 nm (F340/F380) both with 510 nm emission, using a

Nikon Ti2 microscope equipped with a Photometrics Prime 95B camera, followed by data

analysis using Nikon Element software (Dickerson et al., 2018).

Statistics

Two-tailed Student’s t-test and ANOVA was performed with Prism (GraphPad) for quantitative

PCR experiments, glucose tolerance tests, hyperglycemic clamp studies, and secretagogue-

stimulated insulin secretion.

Study approval

The Columbia University Institutional Animal Care and Utilization Committee approved all

experiments.

20 bioRxiv preprint doi: https://doi.org/10.1101/2020.07.04.187864; this version posted July 5, 2020. The copyright holder for this preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under aCC-BY-NC-ND 4.0 International license.

Kuo et al, Hnf4a and FoxO1 in b-cell function

Acknowledgments

We give special thanks to Mitchell A. Lazar (University of Pennsylvania) and Manashree Damle

(Massachusetts General Hospital) for their bioinformatics support and guidance. We also thank

members of the Accili laboratory for discussions, and Thomas Kolar, Ana Flete-Castro, and

Qiong Xu (Columbia University) for technical support. This work was supported by NIH grants

T32DK07328 (to T.K.), K01DK114372 (to T.K.), DK097392 (to D.A.J.), DK115620 (to D.A.J.),

DK64819 (to D.A.), DK63608 (to Columbia University Diabetes Research Center), and JPB

Foundation (to D.A. and Mitchell A. Lazar).

21 bioRxiv preprint doi: https://doi.org/10.1101/2020.07.04.187864; this version posted July 5, 2020. The copyright holder for this preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under aCC-BY-NC-ND 4.0 International license.

Kuo et al, Hnf4a and FoxO1 in b-cell function

Duality of interest

The authors declare that they have no conflict of interest.

22 bioRxiv preprint doi: https://doi.org/10.1101/2020.07.04.187864; this version posted July 5, 2020. The copyright holder for this preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under aCC-BY-NC-ND 4.0 International license.

Kuo et al, Hnf4a and FoxO1 in b-cell function

Author Contributions

Conceptualization, T.K. and D.A.; Methodology, T.K. and D.A.; Investigation, T.K., W.D., Y.M.,

P.K.D.; Formal analysis, T.K., W.D., Y.M., P.K.D., D.A.J., and D.A.; Writing-original draft, T.K.

and D.A.; Writing-review and editing, T.K., W.D., Y.M., P.K.D., D.A.J., and D.A.; Funding

acquisition, T.K., D.A.J., and D.A.

23 bioRxiv preprint doi: https://doi.org/10.1101/2020.07.04.187864; this version posted July 5, 2020. The copyright holder for this preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under aCC-BY-NC-ND 4.0 International license.

Kuo et al, Hnf4a and FoxO1 in b-cell function

References

Abdel-Magid, A.F. (2019). Treatment of Diabetes, Obesity, Dyslipidemia, and Related

Disorders with GPR119 Agonists. ACS Med Chem Lett 10, 14-15.

Asahara, S., Shibutani, Y., Teruyama, K., Inoue, H.Y., Kawada, Y., Etoh, H., Matsuda, T.,

Kimura-Koyanagi, M., Hashimoto, N., Sakahara, M., et al. (2013). Ras-related C3 botulinum

toxin substrate 1 (RAC1) regulates glucose-stimulated insulin secretion via modulation of F-

actin. Diabetologia 56, 1088-1097.

Balamatsias, D., Kong, A.M., Waters, J.E., Sriratana, A., Gurung, R., Bailey, C.G., Rasko, J.E.,

Tiganis, T., Macaulay, S.L., and Mitchell, C.A. (2011). Identification of P-Rex1 as a novel Rac1-

guanine nucleotide exchange factor (GEF) that promotes actin remodeling and GLUT4 protein

trafficking in adipocytes. J Biol Chem 286, 43229-43240.

Boj, S.F., Petrov, D., and Ferrer, J. (2010). Epistasis of transcriptomes reveals synergism

between transcriptional activators Hnf1alpha and Hnf4alpha. PLoS Genet 6, e1000970.

Brouwers, B., de Faudeur, G., Osipovich, A.B., Goyvaerts, L., Lemaire, K., Boesmans, L.,

Cauwelier, E.J., Granvik, M., Pruniau, V.P., Van Lommel, L., et al. (2014). Impaired islet

function in commonly used transgenic mouse lines due to human growth hormone minigene

expression. Cell Metab 20, 979-990.

Buteau, J., Shlien, A., Foisy, S., and Accili, D. (2007). Metabolic diapause in pancreatic beta-

cells expressing a gain-of-function mutant of the forkhead protein Foxo1. J Biol Chem 282, 287-

293.

Chen, F., Sha, M., Wang, Y., Wu, T., Shan, W., Liu, J., Zhou, W., Zhu, Y., Sun, Y., Shi, Y., et

al. (2016). Transcription factor Ets-1 links glucotoxicity to pancreatic beta cell dysfunction

through inhibiting PDX-1 expression in rodent models. Diabetologia 59, 316-324.

24 bioRxiv preprint doi: https://doi.org/10.1101/2020.07.04.187864; this version posted July 5, 2020. The copyright holder for this preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under aCC-BY-NC-ND 4.0 International license.

Kuo et al, Hnf4a and FoxO1 in b-cell function

Chen, Y., Takizawa, N., Crowley, J.L., Oh, S.W., Gatto, C.L., Kambara, T., Sato, O., Li, X.D.,

Ikebe, M., and Luna, E.J. (2003). F-actin and myosin II binding domains in supervillin. J Biol

Chem 278, 46094-46106.

Cinti, F., Bouchi, R., Kim-Muller, J.Y., Ohmura, Y., Sandoval, P.R., Masini, M., Marselli, L.,

Suleiman, M., Ratner, L.E., Marchetti, P., et al. (2016). Evidence of beta-Cell Dedifferentiation in

Human Type 2 Diabetes. J Clin Endocrinol Metab 101, 1044-1054.

Creyghton, M.P., Cheng, A.W., Welstead, G.G., Kooistra, T., Carey, B.W., Steine, E.J., Hanna,

J., Lodato, M.A., Frampton, G.M., Sharp, P.A., et al. (2010). Histone H3K27ac separates active

from poised enhancers and predicts developmental state. Proc Natl Acad Sci U S A 107, 21931-

21936.

Dickerson, M.T., Bogart, A.M., Altman, M.K., Milian, S.C., Jordan, K.L., Dadi, P.K., and

Jacobson, D.A. (2018). Cytokine-mediated changes in K(+) channel activity promotes an

adaptive Ca(2+) response that sustains beta-cell insulin secretion during inflammation. Sci Rep

8, 1158.

Dissanayake, W.C., Sorrenson, B., and Shepherd, P.R. (2018). The role of adherens junction

proteins in the regulation of insulin secretion. Biosci Rep 38.

Dobin, A., Davis, C.A., Schlesinger, F., Drenkow, J., Zaleski, C., Jha, S., Batut, P., Chaisson,

M., and Gingeras, T.R. (2013). STAR: ultrafast universal RNA-seq aligner. Bioinformatics 29,

15-21.

Fajans, S.S., and Bell, G.I. (2011). MODY: history, genetics, pathophysiology, and clinical

decision making. Diabetes Care 34, 1878-1884.

Fan, J., Du, W., Kim-Muller, J.Y., Son, J., Kuo, T., Larrea, D., Garcia, C., Kitamoto, T.,

Kraakman, M.J., Owusu-Ansah, E., et al. (2020). Cyb5r3 links FoxO1-dependent mitochondrial

dysfunction with beta-cell failure. Molecular metabolism 34, 97-111.

Ferrannini, E. (2010). The stunned beta cell: a brief history. Cell metabolism 11, 349-352.

25 bioRxiv preprint doi: https://doi.org/10.1101/2020.07.04.187864; this version posted July 5, 2020. The copyright holder for this preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under aCC-BY-NC-ND 4.0 International license.

Kuo et al, Hnf4a and FoxO1 in b-cell function

Fujitani, Y. (2017). Transcriptional regulation of pancreas development and beta-cell function

[Review]. Endocr J 64, 477-486.

Ganjam, G.K., Dimova, E.Y., Unterman, T.G., and Kietzmann, T. (2009). FoxO1 and HNF-4

are involved in regulation of hepatic glucokinase gene expression by resveratrol. J Biol Chem

284, 30783-30797.

Gardner, D.S., and Tai, E.S. (2012). Clinical features and treatment of maturity onset diabetes

of the young (MODY). Diabetes Metab Syndr Obes 5, 101-108.

Gupta, R.K., Vatamaniuk, M.Z., Lee, C.S., Flaschen, R.C., Fulmer, J.T., Matschinsky, F.M.,

Duncan, S.A., and Kaestner, K.H. (2005). The MODY1 gene HNF-4alpha regulates selected

genes involved in insulin secretion. J Clin Invest 115, 1006-1015.

Hart, A.W., Mella, S., Mendrychowski, J., van Heyningen, V., and Kleinjan, D.A. (2013). The

developmental regulator Pax6 is essential for maintenance of islet cell function in the adult

mouse pancreas. PLoS One 8, e54173.

Herman, W.H., Fajans, S.S., Ortiz, F.J., Smith, M.J., Sturis, J., Bell, G.I., Polonsky, K.S., and

Halter, J.B. (1994). Abnormal insulin secretion, not insulin resistance, is the genetic or primary

defect of MODY in the RW pedigree. Diabetes 43, 40-46.

Herrera, P.L. (2000). Adult insulin- and glucagon-producing cells differentiate from two

independent cell lineages. Development 127, 2317-2322.

Hirota, K., Sakamaki, J., Ishida, J., Shimamoto, Y., Nishihara, S., Kodama, N., Ohta, K.,

Yamamoto, M., Tanimoto, K., and Fukamizu, A. (2008). A combination of HNF-4 and Foxo1 is

required for reciprocal transcriptional regulation of glucokinase and glucose-6-phosphatase

genes in response to fasting and feeding. J Biol Chem 283, 32432-32441.

Kalwat, M.A., and Thurmond, D.C. (2013). Signaling mechanisms of glucose-induced F-actin

remodeling in pancreatic islet beta cells. Exp Mol Med 45, e37.

26 bioRxiv preprint doi: https://doi.org/10.1101/2020.07.04.187864; this version posted July 5, 2020. The copyright holder for this preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under aCC-BY-NC-ND 4.0 International license.

Kuo et al, Hnf4a and FoxO1 in b-cell function

Kim-Muller, J.Y., Fan, J., Kim, Y.J., Lee, S.A., Ishida, E., Blaner, W.S., and Accili, D. (2016a).

Aldehyde dehydrogenase 1a3 defines a subset of failing pancreatic beta cells in diabetic mice.

Nature communications 7, 12631.

Kim-Muller, J.Y., Kim, Y.J., Fan, J., Zhao, S., Banks, A.S., Prentki, M., and Accili, D. (2016b).

FoxO1 Deacetylation Decreases Fatty Acid Oxidation in beta-Cells and Sustains Insulin

Secretion in Diabetes. J Biol Chem 291, 10162-10172.

Kim-Muller, J.Y., Zhao, S., Srivastava, S., Mugabo, Y., Noh, H.L., Kim, Y.R., Madiraju, S.R.,

Ferrante, A.W., Skolnik, E.Y., Prentki, M., et al. (2014). Metabolic inflexibility impairs insulin

secretion and results in MODY-like diabetes in triple FoxO-deficient mice. Cell Metab 20, 593-

602.

Kitamura, T., Kitamura, Y.I., Kobayashi, M., Kikuchi, O., Sasaki, T., Depinho, R.A., and Accili,

D. (2009). Regulation of pancreatic juxtaductal endocrine cell formation by FoxO1. Mol Cell Biol

29, 4417-4430.

Kowluru, A. (2017). Role of G-proteins in islet function in health and diabetes. Diabetes Obes

Metab 19 Suppl 1, 63-75.

Kuo, T., Chen, T.C., Yan, S., Foo, F., Ching, C., McQueen, A., and Wang, J.C. (2014).

Repression of glucocorticoid-stimulated angiopoietin-like 4 gene transcription by insulin. J Lipid

Res 55, 919-928.

Kuo, T., Damle, M., Gonzalez, B.J., Egli, D., Lazar, M.A., and Accili, D. (2019a). Induction of

alpha cell-restricted Gc in dedifferentiating beta cells contributes to stress-induced beta-cell

dysfunction. JCI Insight 5.

Kuo, T., Kim-Muller, J.Y., McGraw, T.E., and Accili, D. (2016). Altered Plasma Profile of

Antioxidant Proteins as an Early Correlate of Pancreatic beta Cell Dysfunction. J Biol Chem

291, 9648-9656.

27 bioRxiv preprint doi: https://doi.org/10.1101/2020.07.04.187864; this version posted July 5, 2020. The copyright holder for this preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under aCC-BY-NC-ND 4.0 International license.

Kuo et al, Hnf4a and FoxO1 in b-cell function

Kuo, T., Kraakman, M.J., Damle, M., Gill, R., Lazar, M.A., and Accili, D. (2019b). Identification

of C2CD4A as a human diabetes susceptibility gene with a role in beta cell insulin secretion.

Proc Natl Acad Sci U S A 116, 20033-20042.

Kuo, T., Lew, M.J., Mayba, O., Harris, C.A., Speed, T.P., and Wang, J.C. (2012). Genome-

wide analysis of glucocorticoid receptor-binding sites in myotubes identifies gene networks

modulating insulin signaling. Proc Natl Acad Sci U S A 109, 11160-11165.

Langlet, F., Haeusler, R.A., Linden, D., Ericson, E., Norris, T., Johansson, A., Cook, J.R.,

Aizawa, K., Wang, L., Buettner, C., et al. (2017). Selective Inhibition of FOXO1

Activator/Repressor Balance Modulates Hepatic Glucose Handling. Cell 171, 824-835 e818.

Lenhard, B., Sandelin, A., and Carninci, P. (2012). Metazoan promoters: emerging

characteristics and insights into transcriptional regulation. Nat Rev Genet 13, 233-245.

Li, H., and Durbin, R. (2009). Fast and accurate short read alignment with Burrows-Wheeler

transform. Bioinformatics 25, 1754-1760.

Liao, Y., Smyth, G.K., and Shi, W. (2014). featureCounts: an efficient general purpose

program for assigning sequence reads to genomic features. Bioinformatics 30, 923-930.

Liu, H.L., Yang, H.Y., Liu, L.X., Chen, Y., Kuang, H.Y., Zhang, H.J., and Yin, H.Q. (2011).

Relationship between down-regulation of HIC1 and PTEN genes and dysfunction of pancreatic

islet cells in diabetic rats. Acta Histochem 113, 340-348.

Love, M.I., Huber, W., and Anders, S. (2014). Moderated estimation of fold change and

dispersion for RNA-seq data with DESeq2. Genome Biol 15, 550.

Luo, Y., He, F., Hu, L., Hai, L., Huang, M., Xu, Z., Zhang, J., Zhou, Z., Liu, F., and Dai, Y.S.

(2014). Transcription factor Ets1 regulates expression of thioredoxin-interacting protein and

inhibits insulin secretion in pancreatic beta-cells. PLoS One 9, e99049.

Miura, A., Yamagata, K., Kakei, M., Hatakeyama, H., Takahashi, N., Fukui, K., Nammo, T.,

Yoneda, K., Inoue, Y., Sladek, F.M., et al. (2006). Hepatocyte nuclear factor-4alpha is essential

for glucose-stimulated insulin secretion by pancreatic beta-cells. J Biol Chem 281, 5246-5257.

28 bioRxiv preprint doi: https://doi.org/10.1101/2020.07.04.187864; this version posted July 5, 2020. The copyright holder for this preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under aCC-BY-NC-ND 4.0 International license.

Kuo et al, Hnf4a and FoxO1 in b-cell function

Orci, L., Gabbay, K.H., and Malaisse, W.J. (1972). Pancreatic beta-cell web: its possible role in

insulin secretion. Science 175, 1128-1130.

Puigserver, P., Rhee, J., Donovan, J., Walkey, C.J., Yoon, J.C., Oriente, F., Kitamura, Y.,

Altomonte, J., Dong, H., Accili, D., et al. (2003). Insulin-regulated hepatic gluconeogenesis

through FOXO1-PGC-1alpha interaction. Nature 423, 550-555.

Robertson, R.P., Harmon, J., Tran, P.O., Tanaka, Y., and Takahashi, H. (2003). Glucose

toxicity in beta-cells: type 2 diabetes, good radicals gone bad, and the glutathione connection.

Diabetes 52, 581-587.

Rouillard, A.D., Gundersen, G.W., Fernandez, N.F., Wang, Z., Monteiro, C.D., McDermott,

M.G., and Ma'ayan, A. (2016). The harmonizome: a collection of processed datasets gathered

to serve and mine knowledge about genes and proteins. Database (Oxford) 2016.

Sander, M., Neubuser, A., Kalamaras, J., Ee, H.C., Martin, G.R., and German, M.S. (1997).

Genetic analysis reveals that PAX6 is required for normal transcription of pancreatic hormone

genes and islet development. Genes Dev 11, 1662-1673.

Snitkin, E.S., and Segre, D. (2011). Epistatic interaction maps relative to multiple metabolic

phenotypes. PLoS Genet 7, e1001294.

Sun, J., Ni, Q., Xie, J., Xu, M., Zhang, J., Kuang, J., Wang, Y., Ning, G., and Wang, Q. (2019).

beta-Cell Dedifferentiation in Patients With T2D With Adequate Glucose Control and

Nondiabetic Chronic Pancreatitis. J Clin Endocrinol Metab 104, 83-94.

Swisa, A., Avrahami, D., Eden, N., Zhang, J., Feleke, E., Dahan, T., Cohen-Tayar, Y.,

Stolovich-Rain, M., Kaestner, K.H., Glaser, B., et al. (2017). PAX6 maintains beta cell identity by

repressing genes of alternative islet cell types. J Clin Invest 127, 230-243.

Talchai, C., Xuan, S., Lin, H.V., Sussel, L., and Accili, D. (2012). Pancreatic beta cell

dedifferentiation as a mechanism of diabetic beta cell failure. Cell 150, 1223-1234.

Talchai, S.C., and Accili, D. (2015). Legacy Effect of Foxo1 in Pancreatic Endocrine

Progenitors on Adult beta-Cell Mass and Function. Diabetes 64, 2868-2879.

29 bioRxiv preprint doi: https://doi.org/10.1101/2020.07.04.187864; this version posted July 5, 2020. The copyright holder for this preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under aCC-BY-NC-ND 4.0 International license.

Kuo et al, Hnf4a and FoxO1 in b-cell function

Wang, Z., Oh, E., and Thurmond, D.C. (2007). Glucose-stimulated Cdc42 signaling is

essential for the second phase of insulin secretion. J Biol Chem 282, 9536-9546.

Whyte, W.A., Orlando, D.A., Hnisz, D., Abraham, B.J., Lin, C.Y., Kagey, M.H., Rahl, P.B., Lee,

T.I., and Young, R.A. (2013). Master transcription factors and mediator establish super-

enhancers at key cell identity genes. Cell 153, 307-319.

Yamagata, K., Furuta, H., Oda, N., Kaisaki, P.J., Menzel, S., Cox, N.J., Fajans, S.S., Signorini,

S., Stoffel, M., and Bell, G.I. (1996). Mutations in the hepatocyte nuclear factor-4alpha gene in

maturity-onset diabetes of the young (MODY1). Nature 384, 458-460.

Zhang, Y., Liu, T., Meyer, C.A., Eeckhoute, J., Johnson, D.S., Bernstein, B.E., Nusbaum, C.,

Myers, R.M., Brown, M., Li, W., et al. (2008). Model-based analysis of ChIP-Seq (MACS).

Genome Biol 9, R137.

Zhong, M., Wu, Y., Ou, W., Huang, L., and Yang, L. (2019). Identification of key genes

involved in type 2 diabetic islet dysfunction: a bioinformatics study. Biosci Rep 39.

30 bioRxiv preprint doi: https://doi.org/10.1101/2020.07.04.187864; this version posted July 5, 2020. The copyright holder for this preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under aCC-BY-NC-ND 4.0 International license.

Kuo et al, Hnf4a and FoxO1 in b-cell function

Figure Legends

Figure 1. Motif analysis of distal enhancers. (A-B) Scatterplots of H3K27ac regions in distal

enhancers (A) and promoters (B) in WT and IKO β-cells isolated under three conditions: diabetic

multiparous IKO (DM) and WT controls (normal glucose tolerance, or NGT); age-matched

nulliparous (12M NP) IKO and WT; and 3-month-old (3M) IKO and WT controls. There are 417

up-regulated (red) and 149 down-regulated (blue) H3K27ac regions in distal enhancers (A) and

31 up-regulated and 15 down-regulated H3K27ac regions in promoters (B) shared among the

three comparisons. (C) Transcription factor motif analysis in differentially acetylated distal

enhancers between DM and NGT. For each transcription factor motif, consensus or partial

consensus sequence is boxed, and p value is shown.

Figure 2. Glucose tolerance tests, fasting and refeeding response, and hyperglycemic

clamps. (A) Intraperitoneal glucose tolerance test in WT (n=11), IKO (n=13), HKO (n=12), and

DKO (n=10) mice. † represents p < 0.05 for IKO vs. DKO, * indicates p < 0.05 for WT vs. IKO,

and ‡ shows p < 0.05 for HKO vs. DKO. (B) Area under the curve (AUC) in A. Error bars

represent SEM, *p < 0.05 by Student’s t test and ANOVA. (C) Glucose levels in WT (n=7), IKO

(n=6), HKO (n=5), and DKO (n=7) mice within 5 min of intraperitoneal glucose injection. (D-E)

Glucose (D) and plasma insulin levels (E) in fasted or refed WT (n=5), IKO (n=3), HKO (n=4),

and DKO (n=5) mice. (F) Glucose infusion rates (GIRs) during hyperglycemic clamps in WT

(n=9), IKO (n=6), HKO (n=9), and DKO (n=5) mice, and (G) quantification of AUC. Error bars

represent SEM, *p < 0.05 by Student’s t test and ANOVA. (H) Circulating insulin levels during

hyperglycemic clamps in WT (n=9), IKO (n=6), HKO (n=9), and DKO (n=5) mice, and (I)

quantification of AUC. Error bars represent SEM, *p < 0.05 by Student’s t test and ANOVA. ND

stands for no difference.

31 bioRxiv preprint doi: https://doi.org/10.1101/2020.07.04.187864; this version posted July 5, 2020. The copyright holder for this preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under aCC-BY-NC-ND 4.0 International license.

Kuo et al, Hnf4a and FoxO1 in b-cell function

Figure 3. Ex vivo insulin secretion in purified islets. (A) High glucose- (16.8 mM, or 16.8G)

stimulated insulin secretion in WT (n=11), IKO (n=10), HKO (n=11), and DKO (n=11) islets. (B)

16.8G and L-arginine- (10 mM, or Arg) stimulated insulin secretion in WT (n=10), IKO (n=7),

HKO (n=6), and DKO (n=8) islets. (C) Glibenclamide- (1 µM) stimulated insulin secretion in WT

(n=21), IKO (n=14), HKO (n=9), and DKO (n=13) islets. (D) PMA- (1 µM) stimulated insulin

secretion in WT (n=18), IKO (n=9), HKO (n=14), and DKO (n=8) islets. (E) Exendin-4- (Ex4, 1

µM) stimulated insulin secretion in WT (n=8), IKO (n=8), and DKO (n=8) islets. (F) GIP- (1 µM)

stimulated insulin secretion in WT (n=8), IKO (n=8), and DKO (n=8) islets. Islets were first

incubated in basal glucose (2.8 mM, or 2.8G), followed by the indicated secretagogue(s).

Results are normalized to total protein contents and duration of treatments. Error bars indicate

SEM, ⋈ represents p < 0.05 when comparing basal glucose- and secretagogue- stimulated

insulin secretion in the same genotype by Student’s t test, while * shows p < 0.05 when

comparing secretagogue-stimulated insulin secretion in different genotypes by Student’s t test

and ANOVA.

Figure 4. Average whole islet calcium response. Islets were incubated for 20 minutes in 2

mM glucose-containing solution and imaged using FURA2 prior to treatment with 14 mM

glucose. (A) Representative Fura-2 AM recordings (F340/F380) of changes in whole-islet

intracellular calcium flux for WT, IKO, HKO, and DKO mice. The secretagogues employed are

indicated on top. (B) Area under the curve (AUC) for different secretagogue-stimulated insulin

secretion in A. Error bars indicate SEM, * represents p <0.05 when comparing to WT by

Student’s t test and ANOVA.

Figure 5. Epistatic modalities between FoxO1 and Hnf4α. (A) Antagonistic epistasis-

regulated glycolytic network. (B) Mixed epistasis for solute carrier lineage. (C) Synergistic

32 bioRxiv preprint doi: https://doi.org/10.1101/2020.07.04.187864; this version posted July 5, 2020. The copyright holder for this preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under aCC-BY-NC-ND 4.0 International license.

Kuo et al, Hnf4a and FoxO1 in b-cell function

epistasis-controlled protocadherins. Heatmaps demonstrate altered gene expression in DKO vs.

IKO or HKO β-cells. Scale bars indicate FPKM.

33 bioRxiv preprint doi: https://doi.org/10.1101/2020.07.04.187864; this version posted July 5, 2020. The copyright holder for this preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under aCC-BY-NC-ND 4.0 International license.

Kuo et al, Hnf4a and FoxO1 in b-cell function

Supplemental Materials

Supplemental Figure S1. Increased occupancy of H3K27 acetylation in DM β-cells. (A-B)

Venn diagram showing the genome-wide overlap between NGT and DM for (A) H3K27ac active

regions and (B) genes with H3K27ac peaks. (C) A representative pie chart of percent

distribution of H3K27ac peaks.

Supplemental Figure S2. Clamp glucose levels and insulin tolerance tests. (A) Average

body weight of 6-month old WT (n=11), IKO (n=13), HKO (n=12), and DKO (n=10) mice. (B-C)

Glucose levels during hyperglycemic clamps. (B) Glucose levels during hyperglycemic clamps

in WT (n=9), IKO (n=6), HKO (n=9), and DKO (n=5) mice, and (C) quantification of AUC of B.

(D) Insulin tolerance test in WT (n=8), IKO (n=5), HKO (n=4), and DKO (n=5) mice.

Supplemental Figure S3. Gene expression profiles for (A) potassium channels and (B) non-β-

cell hormones in WT, IKO, HKO, and DKO β-cells. *p < 0.05, **p < 0.01, ***p < 0.001 to WT.

∆ p < 0.05 for HKO vs. DKO. ∇ p < 0.05 for IKO vs. DKO. n=5 for each genotype.

Supplemental Figure S4. Representative examples of increased H3K27ac levels in DKO β-

cells compared to WT, IKO, or both.

34 bioRxiv preprint doi: https://doi.org/10.1101/2020.07.04.187864; this version posted July 5, 2020. The copyright holder for this preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under aCC-BY-NC-ND 4.0 International license.

Kuo et al, Hnf4a and FoxO1 in b-cell function

List of materials

Manuscript

Figure 1: Motif analysis of distal enhancers.

Figure 2: Glucose tolerance tests, fasting and refeeding response, and hyperglycemic clamps.

Figure 3: Ex vivo insulin secretion in purified islets.

Figure 4. Average whole islet calcium response.

Figure 5. Epistatic modalities between FoxO1 and Hnf4α.

Supplemental Figures

Supplemental Figure S1. Increased occupancy of H3K27 acetylation in DM β-cells.

Supplemental Figure S2. Clamp glucose levels and insulin tolerance tests.

Supplemental Figure S3. Gene expression profiles for potassium channels and non-β-cell

hormones in WT, IKO, HKO, and DKO β-cells.

Supplemental Figure S4. Representative examples of increased H3K27ac levels.

Supplemental Tables

Supplemental Table S1: Differential gene expressions in WT vs. IKO beta cells.

Supplemental Table S2: Differential gene expressions in WT vs. HKO beta cells.

Supplemental Table S3: Differential gene expressions in WT vs. DKO beta cells.

Supplemental Table S4: Differential gene expressions in IKO vs. DKO beta cells.

Supplemental Table S5: Differential gene expressions in HKO vs. DKO beta cells.

Supplemental Table S6: Differential gene expressions in IKO vs. HKO beta cells.

Supplemental Table S7: Concordant RNA and H3K27ac profiles in WT vs. DKO beta cells.

Supplemental Table S8: Concordant RNA and H3K27ac profiles in IKO vs. DKO beta cells.

35 bioRxiv preprint doi: https://doi.org/10.1101/2020.07.04.187864; this version posted July 5, 2020. The copyright holder for this preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under aCC-BY-NC-ND 4.0 International license. Figure 1

A H3K27ac Distal Enhancers 10 10 10 increased increased increased decreased decreased decreased 8 8 8 PK M) PK M) !"

!" 6 !" 6 PK M) 6

!# og2 (F !# !# l og2 (F O l 4 K 4 4 og2 (F O I K I DM DM l 2 2 2 3M 12 M NP 0 0 0

0 2 4 6 8 10 0 2 4 6 8 10 0 2 4 6 8 10 NGT log2(FPKM) 12M NP WT log2(FPKM) 3M WT log2(FPKM) B H3K27ac Promoter (3kb-TSS-3kb) 10 10 10 increased increased increased decreased decreased decreased 8 8 8 PK M) PK M) !"

6 !" !" 6 PK M) 6

og2 (F !# !#

!# l og2 (F O l 4 K 4 4 og2 (F O I K I DM DM l 2 2 2 3M 12 M NP 0 0 0

0 2 4 6 8 10 0 2 4 6 8 10 0 2 4 6 8 10 NGT log2(FPKM) 12M NP WT log2(FPKM) 3M WT log2(FPKM)

C Transcription Factor Motif Analysis in Distal Enhancer Regions

DM vs. NGT Hyper-acetylated motifs Hypo-acetylated motifs

10 343 increased Hnf4 Pax6 116 decreased -21 -12 8 p value 1 x 10 p value 1 x 10

6 Hic FoxO1

DM -19 -11

4 p value 1 x 10 p value 1 x 10

2 Ets

0 p value 1 x 10-17 0 2 4 6 8 10 NGT bioRxiv preprint doi: https://doi.org/10.1101/2020.07.04.187864; this version posted July 5, 2020. The copyright holder for this preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under aCC-BY-NC-ND 4.0 International license. !"#$%&'( )$**+&,&-./+'!"#$%&')(

A B C ND A B C D 400 † ‡ * ND WT * WT 500 60000 50000 300 * * WT 15040 WT 150 IKO * IKO 300 † IKO 400 IKO HKO * 40000 200 30 HKO HKO HKO 100 40000 100 DKO † 30000 300 200 * DKO DKO DKO * 20 20000 100 200 Weight (g) Weight Glucose AUC Glucose (mg/dL) Glucose (mg/dL) 100 50 Glucose AUC 20000 50 Glucose (mg/dL) Glucose (mg/dL) 10000 Glucose (mg/dL) 10 100 0 0 0 30 60 90 120 0 00 0 0 0 WT IKO HKO DKO Time (min) 0 min2 min5 min 0 min2 min5 min 0 min2 min5 min 0 min2 min5 min 0WT 15 IKO 30 HKO 45 DKO 60 0 30 60 90 120 WT IKO HKO DKO 0 15 30 45 60 WT IKO HKO DKO Time (min) Time (min) Time (min)

D ND E ND * * * * WT 200 * WT 2.5 IKO IKO * 2.0 * * HKO 150 HKO

g/L) 1.5 DKO DKO µ 100 1.0 Insulin (

Glucose (mg/dL) 50 0.5

0 0.0

refedrefed 1 h 2 h refedrefed 1 h 2 h refedrefed 1 h 2 h refedrefed 1 h 2 h refedrefed 1 h 2 h refedrefed 1 h 2 h refedrefed 1 h 2 h refedrefed 1 h 2 h fasted 16 h fasted 16 h fasted 16 h fasted 16 h fasted 16 h fasted 16 h fasted 16 h fasted 16 h WT IKO HKO DKO WT IKO HKO DKO

F G ND H I ND 70 0.8 * 100 * ) 8000 1

- * 60 * * * 0.6 80 min 50 6000 • 1 g/L) - 40 µ 60

kg 0.4

• 30 4000 40 GIR AUC Insulin ( 20 0.2 Insulin AUC 2000

GIR (mg 10 20 0 0.0 0 30 60 90 120 0 0 30 60 90 120 0 Time (min) WT IKO HKO DKO Time (min) WT IKO HKO DKO bioRxiv preprint doi: https://doi.org/10.1101/2020.07.04.187864; this version posted July 5, 2020. The copyright holder for this preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under aCC-BY-NC-ND 4.0 International license. !"#$%&'(

High glucose High glucose + arginine ooooo-20200413_GSIS_drugs_arg_to_total_prot A B * ooooo-20200413_GSIS_drugs_16.8glu_to_total_prot* * * 0.6 * 1.5 *

! 1.0 ! ! ! 0.4 0.5 ! 0.4

Insulin ! Insulin 0.2 ! ! 0.2 g per mg protein hr) g per mg protein hr) µ ( µ ( 0.0 0.0

WT, 2.8G WT, 2.8G WT, 16.8GIKO, 2.8G HKO, 2.8G DKO, 2.8G IKO, 2.8G IKO, 16.8G HKO, 16.8G DKO, 16.8G HKO, 2.8G DKO, 2.8G WT, 16.8G+Arg IKO, 16.8G+ArgHKO, 16.8G+ArgDKO, 16.8G+Arg

Sulfonylurea PKC activator C * D ooooo-20200413_GSIS_drugs_gliben_to_total_prot* ooooo-20200413_GSIS_drugs_PMA_to_total_prot* * * * 0.6 2.0 ! ! ! 1.5 0.4 ! ! 1.0 ! Insulin Insulin 0.2 ! 0.5 ! g per mg protein hr) g per mg protein hr) µ µ ( 0.0 ( 0.0

WT, 2.8G IKO, 2.8G WT, 2.8G HKO, 2.8G DKO, 2.8G IKO, 2.8G HKO, 2.8G DKO, 2.8G

WT, 2.8G+Gliben WT, 16.8G+PMA IKO, 2.8G+GlibenHKO, 2.8G+GlibenDKO, 2.8G+Gliben IKO, 16.8G+PMAHKO, 16.8G+PMADKO, 16.8G+PMA

Exendin-4 GIP

E F 3 3 !! ! !!! 2 2

1 1

0.3 0.3 Insulin Insulin 0.2 0.2 g per mg protein hr) 0.1 g per mg protein hr) 0.1 µ µ ( ( 0.0 0.0

WT, 2.8G WT, 2.8G IKO, 2.8G DKO, 2.8G IKO, 2.8G DKO, 2.8G

WT, 16.8G+Ex4 WT, 16.8G+GIP IKO, 16.8G+Ex4DKO, 16.8G+Ex4 IKO, 16.8G+GIPDKO, 16.8G+GIP bioRxiv preprint doi: https://doi.org/10.1101/2020.07.04.187864; this version posted July 5, 2020. The copyright holder for this preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under aCC-BY-NC-ND 4.0 International license. Figure 4

A B 2.3 2 mM glucose 14 mM glucose 45 mM KCl 2 mM glucose 14 mM glucose 45 mM KCl WT (n= 55 islets) 800 1st response 2nd response IKO (n=48 islets) 2.1 WT HKO (n= 36 islets) IKO 700 * DKO (n=40 islets) 1.9 HKO * DKO 1.7 600 *

1.5 AUC 500 * * 1.3 *

(F340/F380)/(F340/380)0 400 1.1 * * Relative normalized calcium inﬂux

0.9 300 0 200 400 600 800 1000 1200 0-360 360-660 660-1080 1080-1350 Time (sec) Time interval (sec) bioRxiv preprint doi: https://doi.org/10.1101/2020.07.04.187864; this version posted July 5, 2020. The copyright holder for this preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under aCC-BY-NC-ND 4.0 International license. Figure 5

! " # !"#$%&"'(#') <'+86 93"85%'(#') =&">?>)822;7&5@&"8(

WT IKO HKO DKO *&+,- ."/0$ *&+,- ."/0$ *&+,- ."/0$ Npy 0 40 300 Sst 500 Pyy 123)&23('( 9&2:#8;)$55'85( 45&#&)$6785'"( 700 WT IKO HKO DKO 450 WT IKO HKO DKO WT IKO HKO DKO WT IKO HKO DKO WT IKO HKO DKO Ppy 0 0 5 0 Gcg 750 Pglyrp1 Slc9a4 Slco1a6 Pcdh12 1050 Pdk4 Slc28a3 Slc29a4 Pcdhgb8 WT IKO HKO DKO Pgm2l1 Slc38a1 Slc9a3r1 Pcdhb7 Aldob Slc9a2 Slc11a2 Pcdhgb5 Eno1b Slc38a3 Slc35d3 Pcdha9 Pgbd1 Slc1a1 Pcdhb10 0.5 Pfkfb3 20 Slc40a1 Slc2a3 Slc6a8 15 Pcdhga2 Pgf Pcdhb8 Slc52a3 Slc12a7 Fbp2 Slc6a15 Pcdh15 Pfkfb4 Slc38a7 Slc25a31 2 Pcdha12 Ldha Slc2a1 Slc4a11 Pcdhgb7 1.0 Pggt1b Slc25a29 WT IKO HKO DKO WT IKO HKO DKO 40 Slc24a3 Pcdhgc4 0 0 Pdk1 Slc36a1 Pcdhga3 Kctd14 Cdh10 Pgbd5 Slc16a3 Cdh12 Slc26a1 Slc1a4 Pcdhb18 Kcnj3 Pkd1 25 Kcnc1 Cdh15 Slc26a11 Slc9a8 Pcdhb13 Pgam1 Kcnq1 Cdh16 Slc39a8 Slc23a2 Pcdhb21 1.5 Pfkp Kcnq2 Cdh18 Slc14a2 Slc22a23 Pcdhga11 2 Pgm2 Kcnh5 Cdh19 60 Slc25a24 Slc16a10 Pcdha10 Cdh20 Pfkl Kcnab2 Slc6a14 Slc7a1 Pcdhga1 Cdh26 0.5 Pdha1 4 Kcnn1 Pgk1 Slc22a17 Pcdhga10 Cdh3 Slc23a4 Kcnh3 Eno2 Slc44a3 Pcdhgb1 Cdhr3 Slc35f1 35 2.0 Kcnf1 Pdhb Slco4a1 Slc30a9 4 Cdhr5 Lbh 0 Kcnip3 Cdh6 80 Slco1a5 Slc43a2 Pcdhgb6 Kctd8 Slco2a1 Cdh13 Slc25a51 Pcdhgb2 Kcnn3 Slc4a4 2 Cdh9 Eno1 Slc35c2 Kcnj12 Cdh17 200 Slc29a2 Pcdhga7 1.0 Tpi1 Slc37a4 Pcdhb14 4 Kcnab3 Cdh23 400 Slc27a3 6 Gapdh Slc48a1 Pcdhb22 Cdh11 Slc7a11 45 Cdh5 Pkm 600 6 Pcdhga5 6 0 Aldoa Slc15a2 50 Kcnb1 Cdhr4 800 Slc12a2 Slc50a1 Pcdhgc3 8 Kctd15 Cdhr2 WT IKO HKO DKO Slc7a5 100 20 Slc2a6 WT IKO HKO DKO Kctd2 Cdh24 Slc41a2 Slc3a2 150 Cdh7 Kcnmb2 40 1.5 Slc35e3 Slc25a5 200 Kctd13 Slc16a9 WT IKO HKO DKO Kctd20 60 Slc35d2 Kcnk16 0 Slc25a15 80 Cdh4 Slc25a53 WT IKO HKO DKO Cdh22 Slc46a1 8 Cdh8 15 Slc41a3 Cdh2 Slc9a3r2 Cdhr1 30 Slc8a1 WT IKO HKO DKO Slc12a5 Slc15a4 70 Cdh1 Slco3a1 85 Slc25a35 WT IKO HKO DKO Slc16a6 Slc7a6 10 bioRxiv preprint doi: https://doi.org/10.1101/2020.07.04.187864; this version posted July 5, 2020. The copyright holder for this preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under aCC-BY-NC-ND 4.0 International license. Supplemental Figure S1

A Active Regions with H3K27ac Peaks B Genes with H3K27ac Peaks C Percent Distribution of H3K27ac Peaks PromoterPromoter (0 (0to -to -3kb) 3kb)16% 16% Transcription TranscriptionTermination Termination Intergenic DownstreamDownstream (0 to 3kb) (0- 14% 3kb) 7% DM DM 7% 5'UTR NGT 3`-UTR 8% 13519 5476 NGT 1542 3’UTR2% 3905 11991 312 2% Exon 9%

Intron 44% bioRxiv preprint doi: https://doi.org/10.1101/2020.07.04.187864; this version posted July 5, 2020. The copyright holder for this preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under aCC-BY-NC-ND 4.0 International license. )$**+&,&-./+'!"#$%&')( !"#$%&'(

D ND E ND * * * * WT 200 * WT 2.5 IKO IKO * 2.0 * * HKO 150 HKO g/L) 1.5 DKO DKO µ 100 1.0 Insulin (

Glucose (mg/dL) 50 0.5

0 0.0

F G ND H I ND 70 0.8 * 100 * ) 8000 1

- * 60 * * * 0.6 80 min 50 6000 • 1 g/L) - 40 µ 60 kg 0.4

• 30 4000 40 GIR AUC Insulin ( 20 0.2 Insulin AUC 2000

Supplemental Figure S3

! !"#$%%&'()*+$,,-.%)/0123

2.0 WT IKO ** 1.5 HKO ** DKO * ** 1.0 ** ** ** *

RNAseq (FPKM) 0.5 * * * 0.0 Kctd14 Kcnj3 Kcnc1 Kcnq1 Kcnq2 Kcnh5 Kcnab2

8 WT IKO ! 6 HKO DKO *** * 4 ! ** ** *** ** *

RNAseq (FPKM) 2 **

0 Kcnn1 Kcnh3 Kcnf1 Kcnip3 Kctd8 Kcnn3 Kcnj12 Kcnab3

100 WT IKO 65 HKO DKO 30 30 ** 20 *

RNAseq (FPKM) * 10 * 0 Kcnb1 Kctd15 Kctd2 Kcnmb2 Kctd13 Kctd20 Kcnk16

" 2",454*-..)+"6(",-%

WT ! 1500 IKO ! 1000 HKO ! DKO 500 ! 50 40 ** RNAseq (FPKM) 30 *** 20 * 10 0 Npy Sst Pyy Ppy Gcg bioRxiv preprint doi: https://doi.org/10.1101/2020.07.04.187864; this version posted July 5, 2020. The copyright holder for this preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under aCC-BY-NC-ND 4.0 International license. Supplemental Figure S4

!"#$%&'%()*+,-.&#)/")0,1) 2#%33')#456&$%()74)89

75,074 75,076 75,078 75,080 kb 143,350 143,352 143,354 143,356 143,358 143,360 143,362 kb !

!" 25 25 0 0 25 25 #$% 0 0 25 25 &$% 0 0 25 25 '$% 0 0

-./#0 1"".23

!"#$%&'%()*+,-.&#)/")0,1) 2#%33')#456&$%()74)!,1

95,160 95,162 95,164 95,166 95,168 kb 95,764 95,766 95,768 95,770 95,772 95,774 95,776 95,778 95,780 95,782 kb ! !" 25 25 0 0 25 25 #$% 0 0 25 25 &$% 0 0 25 25 '$% 0 0

!"#$%&' ()$*+,

!"#$%&'%()*+,-.&#)/")0,1) 2#%33')#456&$%()74)89)&"()!,1

166,628 166,630 166,632 166,634 166,636 kb 5,040 5,042 5,044 5,046 5,048 5,050 kb ! !" 50 20 0 0 #$% 50 20 0 0 &$% 50 20 0 0 '$% 50 20 0 0

4"%56 789: